Tài liệu Báo cáo khoa học: Specific targeting of a DNA-alkylating reagent to mitochondria Synthesis and characterization of [4-((11aS)-7-methoxy-1,2,3,11a-tetrahydro-5H-pyrrolo[2,1-c][1,4]benzodiazepin-5-on-8-oxy)butyl]-triphenylphosphonium iodide doc

Specific targeting of a DNA-alkylating reagent to mitochondria
Synthesis and characterization of [4-((11aS)-7-methoxy-1,2,3,11a-tetrahydro-
5
H
-pyrrolo[2,1-
c
][1,4]benzodiazepin-5-on-8-oxy)butyl]-triphenylphosphonium iodide
Andrew M. James
1
, Frances H. Blaikie
2
, Robin A. J. Smith
2
, Robert N. Lightowlers
3
, Paul M. Smith
3
and Michael P. Murphy
1
1
MRC-Dunn Human Nutrition Unit, Wellcome Trust-MRC Building, Cambridge, UK;
2
Department of Chemistry, University
of Otago, Dunedin, New Zealand;
3
Department of Neurology, Medical School, University of Newcastle upon Tyne, UK
The selective manipulation of the expression and replica-
tion of mitochondrial DNA (mtDNA) within mammalian
cells has proven difficult. In progressing towards this goal
we synthesized a novel mitochondria-targeted DNA-
alkylating reagent. The active alkylating moiety [(11aS)-8-

hydroxy-7-methoxy-1,2,3,11a-tetrahydro-5H-pyrrolo[2,1-c]
[1,4]benzodiazepin-5-one (DC-81)], irreversibly alkylates
guanine bases in DNA (with a preference for AGA tri-
plets), preventing its expression and replication. To target
this compound to mitochondria it was covalently coupled
to the lipophilic triphenylphosphonium (TPP) cation to
form a derivative referred to as mitoDC-81. Incorporation
of this lipophilic cation led to the rapid uptake of
mitoDC-81 by mitochondria, driven by the large mem-
brane potential across the inner membrane. This com-
pound efficiently alkylated isolated supercoiled, relaxed-
circular or linear plasmid DNA and isolated mtDNA.
However mitoDC-81 did not alkylate mtDNA within
isolated mitochondria or cells, even though it accessed the
mitochondrial matrix at concentrations up to 100-fold
higher than those required to alkylate isolated DNA. This
surprising finding suggests that mtDNA within intact
mitochondria may not be accessible to this class of alky-
lating reagent. This inability to alkylate mtDNA in situ
has significant implications for the design of therapies for
mtDNA diseases and for studies on the packaging,
expression and turnover of mtDNA in general.
Keywords: membrane potential; mitochondria; mitochond-
rial DNA; targeting.
Mammalian mitochondrial DNA (mtDNA) encodes 13
polypeptides and the RNA machinery for their transcrip-
tion and translation [1–3]. As these polypeptides are all
components of oxidative phosphorylation complexes,
mtDNA mutations can severely disrupt mitochondrial
function, leading to a number of human diseases for which

there are no effective therapies [2–8]. Possibilities for
treatment, such as the replacement of the defective gene
by gene therapy, are being explored; however, gene therapy
for mtDNA diseases is even more challenging than for
nuclear gene defects because of the problem of delivering
DNA to mitochondria, the difficulty of generating stable
insertion or expression of exogenous DNA within mam-
malian mitochondria, and the large number of mitochon-
dria and mtDNA molecules per cell [9,10]. Even if this
approach is effective, it will not be practical for the majority
of mtDNA diseases that are caused by mtDNA deletions, or
mutations in RNA genes [9,11] until we extend our
knowledge of potential RNA import processes in mamma-
lian mitochondria [12].
Because of these challenges an alternative ÔantigenomicÕ
strategy has been developed as a potential therapy for
mtDNA diseases [13–16]. This approach does not introduce
a functioning copy of the defective gene; instead it utilizes
the following mtDNA properties: mtDNA is present in
patients at high copy number as a mixture of both normal
and mutated molecules; mtDNA diseases are only pheno-
typically expressed above a threshold proportion of mutated
mtDNA; and mtDNA is continually degraded and resyn-
thesized. Consequently, if the proportion of mutated
mtDNA molecules in a patient can be decreased below this
threshold the disease phenotype may be suppressed. This
could be done by selectively enhancing the degradation, or
inhibiting the replication, of mutated mtDNA molecules
without affecting wild-type mtDNA [17]. The potential of
this approach has been demonstrated for the mtDNA

disease neuropathy, ataxia and retinitis pigmentosa
Correspondence to M. P. Murphy, MRC-Dunn Human Nutrition
Unit, Wellcome Trust-MRC Building, Hills Road,
Cambridge CB2 2XY, UK.
Fax: + 44 1223 252905, Tel.: + 44 1223 252900,
E-mail: ,

Abbreviations: DC-81, (11aS)-8-hydroxy-7-methoxy-1,2,3,11a-
tetrahydro-5H-pyrrolo[2,1-c][1,4]benzodiazepin-5-one; DMEM,
Dulbecco’s modified Eagle medium; FCCP, carbonylcyanide-p-
trifluoromethoxy-phenylhydrazone; IBTP, 4-iodobutyltriphenyl-
phosphonium iodide; mitoDC-81, [4-((11aS)-7-methoxy-1,2,3,11a-
tetrahydro-5H-pyrrolo[2,1-c][1,4]benzodiazepin-5-on-8-oxy)butyl]tri-
phenylphosphonium iodide; mtDNA, mitochondrial DNA; NARP,
neuropathy, ataxia and retinitis pigmentosa; PNA, peptide nucleic
acid; TPMP, methyl triphenylphosphonium; TPP, triphenyl-
phosphonium.
(Received 27 February 2003, revised 15 April 2003,
accepted 12 May 2003)
Eur. J. Biochem. 270, 2827–2836 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03660.x
(NARP), by targeting a restriction enzyme to mitochondria
that cleaves at a unique site associated with the NARP point
mutation [18]. While these experiments show that this
approach works in vitro, the targeting of specific restriction
enzymes is difficult to achieve in patients and most mtDNA
diseases are not associated with a unique restriction site.
Therefore other reagents that selectively inhibit the replica-
tion of mutated mtDNA molecules are being developed [16].
Among the most promising are peptide nucleic acids (PNA),
which are stable nucleic acid analogues that bind tightly to

complementary DNA sequences [19], selectively inhibiting
the replication of mutated mtDNA sequences without
disrupting wild-type sequences that differ by only a single
base pair [20]. Furthermore PNAs can be delivered to the
mitochondrial matrix, by attachment of a mitochondrial
protein import sequence [21], or conjugation to a lipophilic
cation [22]. However, even though both approaches
appeared to lead to the accumulation of a PNA
within the mitochondria of cultured cells, neither affected
the proportion of mutated mtDNA in heteroplasmic
myoclonic epilepsy with ragged-red fibres (MERRF) cells
[23].
Accepting that the PNA has been successfully transpor-
ted to the site of mtDNA replication, these negative results
suggestthatdeliveryofaPNAthatbindstoaparticular
sequence to mitochondria is not sufficient, perhaps because
the PNA does not form a complex with the target mtDNA
sequence that is durable enough to inhibit mtDNA repli-
cation or expression. One approach to increase the duration
of PNA binding to DNA is to conjugate it to a DNA-
alkylating reagent so that the PNA becomes covalently
bound to its target sequence. To do this a DNA alkylating
reagent that reacts relatively slowly with DNA is required to
ensure that the alkylation occurs following binding to the
specific sequence by the PNA. As a first step towards this
goal we set out to develop a mitochondria-targeted DNA-
alkylating reagent to determine whether it was possible to
alkylate mtDNA within intact mitochondria and cells. In
addition to facilitating the development of antigenomic
therapies, such a reagent might also be useful in investi-

gating the rate of turnover of mtDNA and in the derivation
of mtDNA-free (q°) clones of the large number of cell lines
where the classical treatment with ethidium bromide is
ineffective [24].
Therefore, our immediate objective was to synthesize a
mitochondria-targeted DNA-alkylating reagent that
would covalently bind to mtDNA, leading to inhibition
of replication and expression (Fig. 1). For the DNA
alkylating reagent we chose the antibiotic (11aS)-8-
hydroxy-7-methoxy-1,2,3,11a-tetrahydro-5H-pyrrolo[2,1-
c][1,4]benzodiazepin-5-one; (DC-81) [1,25]. This DNA-
alkylating reagent binds to the minor groove of double-
stranded DNA and then alkylates guanines at position N-
2 [26], favouring guanines flanked by purines [27–29]. The
covalent attachment of DC-81-like compounds to DNA
causes premature termination of transcription in vitro [29].
To target DC-81 to mitochondria we attached it to a
lipophilic triphenylphosphonium (TPP) cation. These
cations are accumulated into the mitochondrial matrix
several-hundred fold, driven by the large mitochondrial
membrane potential, and addition of a TPP-moiety has
been used to drive the selective uptake of a wide variety of
molecules into mitochondria [23,30–33]. Therefore attach-
ing the DNA alkylating reagent DC-81 to TPP should
give rise to a high local concentration in the vicinity of
mtDNA with the resultant binding and selective alkyla-
tion leading to a depletion of mtDNA in intact cells
(Fig. 1). Here we report the synthesis and characterization
of a novel mitochondria-targeted alkylating reagent and
show that it alkylates DNA in vitro and is taken up by

mitochondria. However, in spite of its substantial import,
it did not alkylate mtDNA in isolated mitochondria or
cells. This unexpected finding has significant implications
for the development of antigenomic therapies for mtDNA
diseases.
Fig. 1. Selective uptake of mitoDC-81 by mitochondria and subsequent
alkylation of mtDNA. The uptake of mitoDC-81 into cells driven by
theplasmamembranepotential(Dw
p
) followed by the further accu-
mulation of mitoDC-81 into mitochondria driven by the mitoch-
ondrial membrane potential (Dw
m
) is illustrated. The Nernst equation
indicates a 10-fold increase in accumulation for every 61.5 mV of
membrane potential. This leads to a millimolar concentration of
mitoDC-81 within mitochondria on incubation of cells with high
nanomolar to micromolar concentrations of mitoDC-81. This high
local concentration of mitoDC-81 within mitochondria could then
lead to the alkylation and inactivation of mtDNA, as indicated in the
figure.
2828 A. M. James et al. (Eur. J. Biochem. 270) Ó FEBS 2003
Materials and methods
Synthesis of [4-((11a
S
)-7-methoxy-1,2,3,11a-tetrahydro-
5
H
-pyrrolo[2,1-
c

][1,4]benzodiazepin-5-on-8-oxy)-
butyl]triphenylphosphonium iodide (mitoDC-81) 2
Sodium hydride (4.8 mg, 0.12 mmol, 60% suspension in
oil) was added to a dry Schlenk tube containing a magnetic
stirrer and held under an argon atmosphere. The sodium
hydride was washed three times with pentane, then dried
in vacuo (0.1 mm Hg). Dimethylformamide (0.1 mL) was
then added and the suspension was stirred for 10 min at
room temperature. A solution of DC-81 1 [34] (24.6 mg,
0.1 mmol) in dimethylformamide (0.2 mL) and tetra-
hydrofuran (0.2 mL) was added dropwise to the reaction
vessel and stirred for 3.75 h, after which time the reaction
mixture was a yellow/brown suspension (Scheme 1). A
solution of (4-iodobutyl)triphenylphosphonium iodide
(44.9 mg, 0.1 mmol) [35] in dimethylformamide (0.2 mL)
was then added dropwise to the ice-cooled reaction mixture,
which was subsequently allowed to warm to room tem-
perature overnight and stirred for a further 2 days at room
temperature to give a light brown suspension. Distilled
water (2 mL) was then carefully added and the mixture was
partitioned with dichloromethane (5 · 2mL). The com-
bined organic layers were dried (MgSO
4
) and evaporated to
dryness in vacuo. The residual oil was dissolved in minimal
dichloromethane, precipitated with excess ether and the
solvent layer decanted. The precipitate was then redissolved
in minimal dichloromethane, precipitated with excess ether
and the solvent layer decanted. This precipitation process
was repeated nine times. The residue was dried under

reduced pressure for 3 h yielding 2 as a mustard coloured
solid (12 mg, 0.014 mmol, 14%).
1
HNMRd 7.6–7.9 (m
(16H, P
+
–ArH and H-11),7.46(1H,s,H-6), 6.49 (1H, s,
H-9),4.1–4.2(2H,m,Ar-O-CH
2
), 3.71 (3H, s, O-CH
3
), 3.4–
4.0(5H,m,P
+
–CH
2
, H-11a and H-3), 1.75–2.4 (8H, m,
O-CH
2
-CH
2
,P
+
–CH
2
-CH
2
, H-1andH-2) p.p.m.,
31
PNMR d 25.65 p.p.m. ESMS found (M

+
) 563.2469
calculated for C
35
H
36
O
3
N
2
P(M
+
) 563.2458.
1
H(300MHz)and
31
P(121MHz)NMRspectrawere
recorded on a Varian Unity 300 spectrometer in chloro-
form-d
1
standardized against the solvent peak or external
85% phosphoric acid, respectively. The high resolution
mass spectrum (ESMS) was obtained by G. Willett, School
of Chemistry, University of New South Wales, Sydney,
Australia.
Isolated DNA incubations
Circular plasmid DNA (5 lg; derived from pYES-PA2,
6.8 kb), HindIII digested k DNA (5 lg),oraladderof
linearmarkerDNA(5lg of a 1-kb ladder, Invitrogen) were
incubated for 6 h at 37 °Cin50lL20m

M
Tris/HCl
pH 8.0, 1 m
M
EDTA (TE) supplemented with 500 l
M
mitoDC-81, 500 l
M
methyl triphenylphosphonium (TPMP)
or ethanol carrier. DNA was then precipitated with 1 vol.
0.6
M
NaOAc, 20 m
M
EDTA followed by 2 vols cold
ethanol. After centrifugation (13000 g for 30 min at 4 °C)
the pellet was resuspended in 50 lL10m
M
Tris/HCl
pH 8.0, 0.1 m
M
EDTA and 1 lgwasrunat60 Vona0.6%
agarose/ethidium bromide gel. The DNA was electrotrans-
ferred to a positive nylon membrane (Hybond N+,
Amersham Pharmacia Biotech), treated with 0.4
M
NaOH,
and fixed by UV irradiation. The membrane was then
blocked overnight with 1% (w/v) milk powder, 0.05% (v/v)
Tween-20 in TBS (TBST) before incubation for 1 h with a

1 : 3000 dilution of rabbit anti-TPP serum in 0.1% (w/v)
milk powder/TBST. The membrane was washed three times
with TBST then incubated for 30 min with a 1 : 5000
dilution of anti-rabbit I
g
G conjugated to horseradish
peroxidase (Sigma). The membrane was washed five times
with TBST and antibody binding to TPP moieties was
visualized by enhanced chemiluminescence (Amersham
Pharmacia Biotech).
MtDNA was prepared from isolated rat liver mitochon-
dria (40 mg protein) using a plasmid spin miniprep kit
(Qiagen) at a ratio of 5 mg mitochondrial protein per
column. The initial alkaline lysis was shortened to  30 s
due to the presence of alkali-labile ribonucleotides in
mtDNA [36]. The eluted mtDNA was pooled, precipitated
with NaOAc/ethanol, and the pellet resuspended in 15 lL
TE. When separated by agarose gel electrophoresis, isolated
mtDNA appeared as three species with apparent sizes of 9,
16 and  40 kb compared to linear markers. These corres-
pond to supercoiled, linear and relaxed-circular mtDNA
forms, respectively. Upon digestion with the single-cutter
ClaI, these species resolved into the linear form, migrating at
16 kb. Southern blotting showed that supercoiled and
relaxed-circular DNA transferred very poorly compared
with linear DNA (data not shown), hence mtDNA was
routinely cut with ClaI prior to electrophoresis. MtDNA
was incubated for 6 h at 37 °C, digested with ClaI and
electrophoresed at 14 V on a 0.4% agarose/ethidium
bromide gel. The DNA was then electrotransferred to a

positive nylon membrane and TPP moieties were detected
using anti-TPP serum as above.
Mitochondrial incubations
Rat liver mitochondria were prepared by homogenization
followed by differential centrifugation [37]. Protein concen-
tration was determined using the biuret assay with BSA as
standard [38]. The uptake of mitoDC-81 by mitochondria
was measured at 30 °C using a mitoDC-81-sensitive ion-
Scheme 1. [4-((11aS)-7-methoxy-1,2,3,11a-tetrahydro-5H-pyrrolo[2,1-
c][1,4]benzodiazepin-5-on-8-oxy)butyl]triphenylphosphonium iodide (2;
mitoDC-81).
Ó FEBS 2003 Mitochondrial DNA alkylation (Eur. J. Biochem. 270) 2829
selective electrode suspended in a stirred chamber open to
the atmosphere [39,40]. To calibrate the electrode, five
stepwise additions of 1 l
M
mitoDC-81 were made to 3 mL
250 m
M
sucrose, 5 m
M
Tris/HCl pH 7.4, 1 m
M
EGTA,
containing isolated rat liver mitochondria (1 mg pro-
teinÆmL
)1
). The mitochondria were energized with 10 m
M
succinate and the membrane potential was dissipated

with 1 l
M
carbonylcyanide-p-trifluoromethoxy-phenyl-
hydrazone (FCCP). To measure alkylation of mtDNA,
mitochondria (1 mgÆmL
)1
protein) were suspended in two
100 mL conical flasks each containing 10 mL 250 m
M
sucrose, 5 m
M
Tris/HCl pH 7.4, 1 m
M
EGTA, 1 m
M
EDTA, 10 m
M
succinate supplemented with 5 l
M
mitoDC-81. The flasks were incubated for 1 h at 30 °Cin
a shaking water bath, after which time the mitochondria
were pelleted by centrifugation (10 000 g for 10 min) and
washed with TE. The mtDNA was isolated as above,
resuspended in 15 lL TE, cut with ClaI and separated at
14 V through a 0.4% agarose/ethidium bromide gel. The
DNA was electrotransferred to a positive nylon membrane
and TPP moieties were detected using anti-TPP serum as
above. Respiration rate measurements were carried out in
the stirred and thermostatted chamber of a Clark-type
oxygen electrode (Rank Brothers, Bottisham, UK).

Cell incubations
Human 143B osteosarcoma cells were cultured in Dulbecco’s
modified Eagle medium (DMEM) supplemented with 10%
(v/v) foetal bovine serum, glucose (4.5 gÆL
)1
), streptomycin
sulfate (100 mgÆL
)1
), penicillin G (100 000 UÆL
)1
), uridine
(50 mgÆL
)1
) and pyruvate (100 mgÆL
)1
) unless otherwise
stated. Cells were grown in plastic flasks or on glass
coverslips at 37 °C in a humidified atmosphere of 5%
CO
2
/95% air until confluent.
For confocal microscopy, cells were grown overnight on
glass coverslips then treated for 6 h with 500 n
M
(4-iodo-
butyl)triphenylphosphonium (IBTP), 500 n
M
TPMP, or
500 n
M

mitoDC-81. Coverslips were washed three times in
NaCl/P
i
and then fixed with 4% (w/v) paraformaldehyde
for 30 min at room temperature. They were washed four
more times, and permeabilized and blocked with 10% (v/v)
foetal bovine serum, 0.1% (v/v) Triton X-100, NaCl/P
i
for
10 min at 4 °C before being shaken overnight at 4 °Cwitha
1 : 500 dilution of rabbit anti-TPP serum and a 1 : 100
dilution of an anti-(cytochrome oxidase subunit 1) mAb
(Molecular Probes) in 10% (v/v) foetal bovine serum,
0.05% (v/v) Tween 20, NaCl/P
i
. They were washed three
more times and incubated for 30 min at 4 °C with a 1 : 200
dilution of anti-mouse IgG AlexaFluor 546 (Molecular
Probes) and a 1 : 500 dilution of anti-rabbit IgG-Oregon
Green in 10% (v/v) foetal bovine serum, 0.05% (v/v) Tween
20. After washing a final five times with NaCl/P
i
,the
coverslips were mounted on slides and visualized using a
Nikon Eclipse E800 confocal microscope and a Nikon · 60
(numerical aperture 1.4) oil immersion Plan Apochromat
objective. Oregon green was illuminated with the 488 nm
line of an argon laser and fluorescence detected using a
HQ515/30 filter. The argon laser was then turned off and
Alexafluor 546 was illuminated with a 543-nm neon laser

and fluorescence detected using a 570LP filter. The green
and red channels were acquired using Biorad Lasersharp
2000 and subsequently merged to indicate colocalization.
The laser intensity, iris and gain settings were identical for
all images shown.
To assess whether mitoDC-81 could completely deplete
mtDNA and thereby create q° cells, 143B cells were grown
in DMEM/foetal bovine serum supplemented with
50 mgÆL
)1
uridine and passaged twice per week in the
presence of either 500 n
M
mitoDC-81 or 500 n
M
TPMP for
a period of 17 days [24]. Cells were cloned by dilution and
after a further 28 days their rate of oxygen consumption
respiring on succinate was measured after digitonin-perme-
abilization [41].
Results
Synthesis of mitoDC-81
The compound of interest, mitoDC-81 2, was prepared by
reaction of the phenoxide anion derived from 1 with IBTP
(Scheme 1). This approach was found to be more effective
than creating a bromoalkyl side chain on 1 followed by
reaction with triphenylphosphine. The product 2 (Scheme 1)
was isolated and characterized as described for previously
synthesized complex triphenylphosphonium salts [30,31].
Accumulation of mitoDC-81 by isolated mitochondria

An ion-selective electrode sensitive to mitoDC-81 was
constructed in order to determine whether mitoDC-81 was
accumulated by energized mitochondria (Fig. 2). Mito-
chondria were added to the incubation chamber and the
response of the electrode was then calibrated by sequential
additions of 1 l
M
mitoDC-81, which showed the expected
logarithmic response of the electrode to cation concentra-
Fig. 2. Uptake of mitoDC-81 by energized mitochondria. An ion-
selective electrode was used to measure the uptake of mitoDC-81 into
rat liver mitochondria. To calibrate the electrode five stepwise addi-
tions of 1 l
M
mitoDC-81 were made to a suspension of mitochondria
in a stirred incubation chamber. The addition of the respiratory sub-
strate succinate generated a membrane potential, which caused
mitoDC-81 to be sequestered inside the mitochondria, and this accu-
mulation was reversed by addition of the uncoupler FCCP.
2830 A. M. James et al. (Eur. J. Biochem. 270) Ó FEBS 2003
tion (Fig. 2). When mitochondria were energized by addi-
tion of the respiratory substrate succinate, the concentration
of mitoDC-81 in the incubation chamber decreased to
about 2.5 l
M
, due to sequestration of mitoDC-81 within the
mitochondrial matrix. This corresponds to uptake of about
2.5 nmol mitoDC-81Æmg mitochondrial protein
)1
.Asthe

mitochondrial volume is about 0.5–0.9 lLÆmg protein
)1
under these conditions [42–44] this corresponds to an
intramitochondrial mitoDC-81 concentration of 2.5–5 m
M
,
although the steady-state free matrix concentration is likely
to be about 60–80% lower than this due to binding to
the matrix-facing surface of the inner membrane [45]. When
the membrane potential was dissipated by addition of the
uncoupler FCCP, mitoDC-81 was rapidly released from the
mitochondria (Fig. 2). Therefore mitoDC-81 is accumu-
lated about a thousand-fold within energized mitochondria,
driven by the membrane potential.
Alkylation of isolated DNA by mitoDC-81
To determine if the conjugation of a lipophilic cation to
DC-81 disrupted its ability to alkylate DNA, we determined
whether mitoDC-81 could alkylate DNA in vitro.Todothis
we incubated 500 l
M
mitoDC-81 with linearized DNA
molecules of different sizes (i.e. a standard DNA ÔladderÕ
used as a molecular weight marker). After incubation the
DNA was separated by electrophoresis, electrotransferred
to a nylon membrane and probed for mitoDC-81 covalently
bound to the DNA by immunoblotting using anti-TPP
serum [35] (Fig. 3A,B). This showed considerable alkylation
of the DNA ladder by mitoDC-81 (Fig. 3B, lane 1).
Analysis of the gel prior to electrotransfer showed that the
extent of alkylation was proportional to the amount of

DNA present and was independent of the size of the DNA
fragment (Fig. 3A, lane 1). As mtDNA in vivo is negatively
supercoiled it was also important to compare the ability of
mitoDC-81 to alkylate linear, relaxed-circular, and super-
coiled DNA. Therefore in parallel experiments we investi-
gated the ability of 500 l
M
mitoDC-81 to alkylate a 6.8-kb
plasmid that was present in all three conformations
(Fig. 3A,B). Analysis of the agarose gel prior to electro-
transfer showed three forms, supercoiled, linear and
relaxed-circular plasmid DNA, in order of increasing
apparent molecular weight (Fig. 3A). Comparison of the
ethidium fluorescence (Fig. 3A) and the amount of DNA
alkylation (Fig. 3B) showed that all three plasmid species
were alkylated by mitoDC-81. Incubation with TPMP,
which lacks the DNA-alkylating moiety but is recognized by
the anti-TPP serum [35], showed that the DC-81 moiety was
essential for DNA-labelling (Fig. 3B, lane 2). The concen-
tration of mitoDC-81 used and the time taken for DNA
alkylation are in agreement with the conditions used to react
DC-81 related compounds and isolated DNA [46].
To confirm that mitoDC-81 could alkylate mtDNA, we
isolated mtDNA from rat liver mitochondria and incubated
it with 500 l
M
mitoDC-81 (isolated mtDNA; Fig. 3C,D).
This showed extensive alkylation of the isolated mtDNA
by mitoDC-81. The higher molecular weight band in the
isolated mtDNA is relaxed-circular mtDNA. Its absence

from the immunoblot was due to poor transfer from the gel
to the membrane, as was confirmed by Southern blotting.
The presence of relaxed-circular mtDNA after digestion
with ClaI may indicate alkylation of the restriction site as
untreated mtDNA, or mtDNA isolated from mitoDC-81-
treated mitochondria, was always fully linearized by ClaI
under identical conditions. MtDNA that had been ran-
domly broken during isolation or incubation with mitoDC-
81 was also extensively alkylated and migrated as a smear
when subsequently cut with ClaI( 4–8 kb; data not
shown).
We conclude that conjugation of the lipophilic cation to
DC-81 does not disrupt its ability to alkylate DNA and high
micromolar concentrations of mitoDC-81 are sufficient to
alkylate linear, circular and supercoiled DNA. As these
concentrations of mitoDC-81 are easily achieved within
the mitochondrial matrix on incubation with mitoDC-81
(Fig. 2), mitoDC-81 should also alkylate mtDNA within
mitochondria and cells.
MitoDC-81 does not alkylate mtDNA in isolated
mitochondria
Having ascertained that mitoDC-81 was sequestered by
isolated mitochondria and that it could covalently modify
Fig. 3. MitoDC-81 alkylates linear, relaxed-circular and supercoiled
DNA in vitro. A 1-kb ladder of linear DNA (1 lgÆmL
)1
) or plasmid
DNA (1 lgÆmL
)1
; 6.8 kb) was incubated for 6 h at 37 °Cwith500 l

M
mitoDC-81, 500 l
M
TPMP or ethanol carrier (A and B). MtDNA was
isolated from rat liver mitochondria (40 mg protein) and incubated
with 500 l
M
mitoDC-81 for 6 h and then cut with ClaI (C and D).
DNA was separated by electrophoresis on an agarose gel and the
ethidium bromide fluorescence recorded to quantify DNA (A and C).
The DNA was then electrotransferred to a nylon membrane and the
presence of TPP moieties covalently attached to the DNA was visu-
alized by immunoblotting using anti-TPP serum (B and D).
Ó FEBS 2003 Mitochondrial DNA alkylation (Eur. J. Biochem. 270) 2831
linear, relaxed-circular and supercoiled DNA in vitro,we
next tested whether it could alkylate mtDNA in situ within
intact, isolated mitochondria. Incubations of 6 h with a
concentration of 500 l
M
mitoDC-81 led to extensive
alkylation of isolated DNA (Fig. 3), but such long incuba-
tion times and high concentrations of mitoDC-81 cannot
be used with isolated mitochondria due to uncoupling and
loss of mitochondrial integrity. Isolated mitochondria can
beincubatedwithamaximumof 5 l
M
mitoDC-81
without uncoupling, but as this concentration leads to an
intramitochondrial concentration of 3–5 m
M

(Fig. 2), the
concentration of mitoDC-81 should not be a limiting
factor. Respiration rate measurements indicated that
isolated mitochondria incubated at 30 °C for 1 h could
still be stimulated with uncoupler, but longer incubations
led to the loss of mitochondrial coupling (data not shown).
Therefore it was important to determine if a 1-h incubation
was sufficient to lead to detectable DNA labelling by
mitoDC-81. To do this, isolated plasmid DNA was
incubated with various concentrations of mitoDC-81 for
1 h (Fig. 4A,B). Extensive alkylation of the DNA was
noted after 1 h incubation at concentrations as low as
50 l
M
mitoDC-81 (Fig. 4A,B). That the concentration of
mtDNA within mitochondria is also unlikely to be limiting
under these conditions is supported by the following rough
calculations. There are about 18.7 · 10
9
mtDNA mole-
culesÆmg protein
)1
in rat liver mitochondria [47], giving about
3.05 · 10
14
bpÆmg protein
)1
, or 335 ng DNAÆmg protein
)1
.

Under these conditions the mitochondrial matrix volume
is about 0.5–0.9 lLÆmg protein
)1
[42–44], so the effective
mtDNA concentration within the matrix is 372–670 lg
DNAÆmL
)1
, significantly higher than the concentration of
plasmid DNA (100 lgÆmL
)1
)usedinFig.4AandB.
Consequently, a 1-h incubation of isolated mitochondria
with 5 l
M
mitoDC-81 should lead to detectable alkylation of
mtDNA within mitochondria. However, when energized
mitochondria were incubated with mitoDC-81 for 1 h and
the mtDNA isolated, there was no DNA alkylation
(Fig. 4C,D).
In summary, loading mitochondria with sufficient
mitoDC-81 to alkylate isolated mtDNA does not detectably
label mtDNA within mitochondria.
MitoDC-81 does not alkylate mtDNA within intact cells
It was unexpected that mitoDC-81 did not alkylate mtDNA
within isolated mitochondria, even though a concentration
of mitoDC-81 50–100-fold higher than that required to
alkylate isolated plasmid DNA had accumulated. One
possibility is that a longer exposure to mitoDC-81 than was
possible with isolated mitochondria could lead to mtDNA
labelling. Substantially longer incubation times are possible

with cultured cells, which can be incubated with mitoDC-81
indefinitely without disrupting their mitochondrial mem-
brane potential. Therefore we next investigated whether
mitoDC-81 alkylated mtDNA within intact cells. Cells
couldbeincubatedwithupto500n
M
mitoDC-81 indefin-
itely, although higher concentrations (1–5 l
M
)weretoxic
over 2–24 h. Using 500 n
M
mitoDC-81 will generate an
ample intramitochondrial concentration of mitoDC-81
because mitoDC-81 will be accumulated within the cyto-
plasm driven by the plasma membrane potential and then
be further accumulated within the mitochondria due to the
mitochondrial membrane potential [11]. From the known
plasma and mitochondrial membrane potentials and the cell
and mitochondrial volumes of 143B cells [48] we estimate an
intramitochondrial concentration of  450 l
M
mitoDC-81
for 143B cells incubated with 500 n
M
mitoDC-81.
To see if long-term incubation with mitoDC-81 did result
in alkylation of mtDNA, cells were incubated with 500 n
M
mitoDC-81, IBTP, or TPMP for 24 h and probed using

antiserum against TPP and confocal microscopy to visualize
any mitoDC-81 bound to mtDNA in situ. IBTP was used as
a positive control as it is taken up by mitochondria in cells in
Fig. 4. MitoDC-81 does not alkylate mtDNA in situ within isolated
mitochondria. In (A) and (B) the concentration dependence of alkyla-
tion of plasmid DNA by mitoDC-81 over a 1-h incubation was
determined. Plasmid DNA (100 lgÆmL
)1
; 6.8 kb) was incubated for
1hat30°C with 0–1000 l
M
mitoDC-81. In (C) and (D) rat liver
mitochondria(1mgproteinÆmL
)1
) were incubated for 1 h at 30 °C
with 5 l
M
mitoDC-81 after which time mtDNA was isolated and cut
with ClaI. All DNA samples were separated on an agarose/ethidium
bromide gel and the fluorescence recorded (A and C). The DNA was
then transferred to a nylon membrane and TPP moieties covalently
attached to the DNA were visualized using anti-TPP serum (B and D).
The mitochondrial incubation with mitoDC-81 was repeated twice as
described here with identical results. In addition, several similar incu-
bations of mitochondria with mitoDC-81 but using a variety of dif-
ferent methods to prepare it for electrophoresis, also showed no
evidence for DNA alkylation by mitoDC-81. A 1-kb linear DNA
ladder (1 lg) and HindIII-digested k DNA (1 lg), which had been
incubated for 6 h with 500 l
M

mitoDC-81, were included as positive
controls and molecular mass markers.
2832 A. M. James et al. (Eur. J. Biochem. 270) Ó FEBS 2003
the same way as mitoDC-81, and once there it binds slowly
but irreversibly to protein thiols, enabling it to be detected
by anti-TPP serum [35]. If mitoDC-81 binds to mtDNA it
will also be retained within mitochondria after fixation and
should be visible by immunocytochemistry in the same way
as IBTP. To see if this was the case, cells were incubated
with 500 n
M
IBTP,TPMPormitoDC-81for24h,then
fixed and dual-labelled with antibodies to the mitochondrial
enzyme cytochrome oxidase (red) and to TPP (green;
Fig. 5). The images were then merged, so yellow indicates
colocalization of the TPP and cytochrome oxidase, while
red indicates that only the cytochrome oxidase was detected
and that there was no TPP immunoreactivity. This experi-
ment showed that IBTP colocalized with cytochrome
oxidase due to its accumulation by mitochondria and
subsequent irreversible reaction with protein thiols
(Fig. 5A). Although TPMP will also accumulate within
mitochondria, it did not generate any TPP-labelling as it is
not covalently bound inside the matrix and is lost on
fixation (Fig. 5B). Confocal microscopy of cells incubated
with mitoDC-81 (Fig. 5C) gave a pattern of fluorescence
that resembled that of TPMP, rather than IBTP. This
suggests that mitoDC-81 is not covalently bound to
mtDNA within mitochondria. Incubations from 2 h to
3 days with mitoDC-81 concentrations ranging from

100 n
M
to 5 l
M
gave similar negative results (data not
shown).
It is possible that the amount of mitoDC-81-modified
mtDNA present is below the detection limit for immuno-
cytochemistry. However this seems unlikely because TPP
covalently bound to mitochondrial macromolecules was
very easy to detect by confocal microscopy after incubation
of cells for 30 min with 100 n
M
IBTP (data not shown).
In addition, mitochondrial respiratory complexes such as
cytochrome oxidase, present at  200 pmolÆmg protein
)1
[49] are easily detected by these procedures, as are other
mitochondrial antigens such as those involved in apoptosis
(e.g. Smac [50]), which are probably present at a fraction of
the content of respiratory complexes. Consequently the limit
for the detection of mitochondrial antigens using confocal
microscopy is likely to be in the low pmolÆmg protein
)1
range. There are  18.7 · 10
9
mtDNA moleculesÆmg mito-
chondrial protein
)1
[47] or about 6.1 · 10

14
basesÆmg
protein
)1
, 19.3% of which are guanines corresponding to
1.18 · 10
14
guanine residuesÆmg protein
)1
or about
200 pmol GÆmg protein
)1
. Therefore we would predict that
mitoDC-81alkylationofafewpercentoftheavailable
mtDNA guanine residues should have been detectable by
confocal microscopy after 24 h. Even so, we cannot entirely
exclude the possibility that mitoDC-81 did alkylate a small
proportion of the guanine residues in mtDNA, but that the
amount alkylated was below the threshold for detection, or
that alkylated mtDNA is rapidly degraded or repaired. In
summary, we found no evidence for alkylation of mtDNA
by mitoDC-81 within cells.
Long-term incubation with mitoDC81 did not impair
cellular respiration
For mitoDC-81 to be of potential use in preventing
mtDNA replication it might require only a few molecules
of mitoDC-81 bound per mtDNA molecule. Thus, even
though we could not directly detect alkylation of mtDNA
by mitoDC-81, it remained possible that mitoDC-81 was
bound to mtDNA, but at concentrations below the

detection limit of immunoblotting or confocal microscopy.
If that was the case then long-term incubation of cells with
mitoDC-81 should disrupt mtDNA replication, depleting
mtDNA and ultimately leading to the production of cells
that entirely lacked mtDNA (q°). Therefore we cultured
143B cells with mitoDC-81 for long time periods and
attempted to isolate q° clones. 143B osteosarcoma cells
were chosen because it is easy to deplete these cells of
mtDNA and q° clones derived from 143B cells are robust
[51]. For these experiments the culture medium was also
supplemented with uridine and pyruvate, as these are
required for the growth and survival of any q° cells that
arise [51–53]. After 2.5 weeks of culture ( 15 cell
divisions) with mitoDC-81, clones were isolated and
allowed to recover in the absence of mitoDC-81 before
mitochondrial oxygen consumption was measured. None
of the individual clones (n ¼ 24) that were analysed had
impaired mitochondrial respiration. This lack of detection
of q° clones was not due to recovery and expansion of a
residual population of nonalkylated mtDNA on removal
Fig. 5. MitoDC-81 does not bind to mtDNA in cultured cells. 143B cells
were cultured at 37 °C for 24 h in the presence of 500 n
M
IBTP (A),
500 n
M
TPMP (B) or 500 n
M
mitoDC-81 (C). After fixation mito-
chondria (red) and TPP moieties (green) were visualized using an anti-

(cytochrome oxidase) mAb and anti-TPP serum, respectively.
Ó FEBS 2003 Mitochondrial DNA alkylation (Eur. J. Biochem. 270) 2833
of mitoDC-81 as a bulk culture of 143B cells grown in
500 n
M
mitoDC-81 for 6 weeks had normal rates of
mitochondrial respiration (data not shown). Concentra-
tions of mitoDC-81 >500 n
M
were too toxic for long-term
culture and 500 n
M
mitoDC-81 slowed cell growth slightly
relative to 500 n
M
TPMP or no additions (data not
shown), indicating that the DC-81 moiety of mitoDC-81
was affecting the cells. Interestingly, this growth inhibition
was not related to effects on mtDNA as 500 n
M
mitoDC-
81 completely prevented the growth of, although did not
necessarily kill, a previously established 143B-derived q°
cell line (data not shown). This effect was specific to
mitoDC-81 and was not due to nonspecific disruption by
the lipophilic cation as q° cells cultured with 500 n
M
TPMP grew at the same rate as control incubations (data
not shown). One possible interpretation is that the higher
mitochondrial membrane potential of 143B cells causes

their mitochondria to sequester mitoDC-81 preventing the
toxic reactions in the cytosol or nucleus that disrupt
growth in q° cells. However, as the q° cells still survived
for several days in the presence of mitoDC-81 this would
not have led to the elimination of any q° cells that arose
during incubation with mitoDC-81. In summary, we found
no evidence for the depletion of mtDNA on long-term
incubation of cells with mitoDC-81.
Discussion
We have synthesized a mitochondria-targeted compound
that has the ability to alkylate isolated DNA. This molecule,
mitoDC-81 is taken up by energized mitochondria and
accumulates within the mitochondrial matrix  1000-fold
driven by the membrane potential. Consequently incubation
of energized mitochondria with micromolar concentrations
of mitoDC-81 leads to millimolar concentrations within the
mitochondrial matrix. MitoDC-81 alkylated isolated
mtDNA, and reacted with supercoiled, relaxed-circular or
linear plasmid DNA. Extensive binding was detectable after
a 1-h incubation of isolated DNA with 50 l
M
mitoDC-81.
However, incubation of mitochondria with mitoDC-81 for
a similar duration under conditions where the concentration
within the mitochondrial matrix was estimated to be 50–
100-fold greater did not lead to detectable binding of
mitoDC-81 to mtDNA. This was extended to cells where
incubation with mitoDC-81 also failed to yield detectable
alkylation of mtDNA by confocal microscopy, even though
the mitochondrial uptake of mitoDC-81 was predicted to be

ample to alkylate isolated mtDNA, and despite the fact that
with cells it was possible to incubate for far longer periods
(24 h), than was possible with isolated mitochondria.
Finally, alkylation of mtDNA by mitoDC-81 would be
expected to disrupt mitochondrial biogenesis and lead to
depletion of mtDNA, but there was no generation of q° or
respiratory-deficient cells on long-term incubation with
mitoDC-81. Therefore we found no evidence that mitoDC-
81 alkylates mtDNA in mitochondria or cells.
The reasons for the lack of alkylation of mtDNA
within mitochondria by mitoDC-81 are unclear. The local
concentrations of mitoDC-81 and DNA, and the duration
of the experiments were ample to alkylate isolated DNA.
One possibility is that mitoDC-81 does alkylate mtDNA
within mitochondria, but this modified DNA is then
rapidly degraded. This seems unlikely, as the amount of
mtDNA isolated from treated and untreated mitochon-
dria was similar. In addition, such a scenario would have
been expected to readily generate q° clones, which were
not found. Alternatively, upon alkylation the mtDNA, or
the mitochondria themselves, may have become difficult
to isolate. However, the similar yields of mtDNA from
isolated mitochondria treated with mitoDC-81 again
make this explanation unlikely. Furthermore, the lack of
labelling of mtDNA by mitoDC-81 in intact cells, as
explored by immunocytochemistry, makes it unlikely that
mtDNA was alkylated in situ, but was then selectively lost
in subsequent manipulations during which unmodified
mtDNA was retained. Alternatively, mitoDC-81 may
react with RNA, nucleotides, nucleosides or other biolo-

gical molecules. Such reactions could inactivate a large
proportion of the mitoDC-81 taken up by mitochondria,
thus preventing its reaction with mtDNA. However, this
would require that these other mitoDC-81 targets are lost
upon fixation, as the adducts were not seen by confocal
microscopy. It is possible that the binding of mitoDC81
to membranes and proteins within mitochondria prevents
its reactivity with mtDNA, however, this binding is
reversible, leading to a dynamic equlibrium between
bound and free compound. The protein thiol reagent
IBTP reacts extensively with matrix proteins, even though
it is of similar hydrophobicity to mitoDC81 and will
therefore be membrane-bound to about the same extent
[35]. Consequently it is unlikely that the binding of
mitoDC81 to mitochondrial components eliminates its
reactivity with mtDNA. We also cannot exclude the
possibility that mitoDC-81 does alkylate mtDNA, but
that the altered bases are rapidly excised by very rapid
and effective endogenous repair processes that are active
in isolated mitochondria. While artefactual explanations
for the lack of mitoDC81 binding to mtDNA in situ
cannot be entirely eliminated, the most likely interpret-
ation of the absence of mtDNA alkylation is that even
though the mitoDC-81 was accumulated by mitochondria
at high concentrations it could not react with mtDNA.
Possible reasons for this could be that the mtDNA is not
accessible to this alkylating reagent due to its interaction
with the inner membrane or the abundance of nucleoid
proteins. The latter have been reported under certain
conditions to entirely wrap mtDNA and cause a marked

increase in nuclease resistance in vitro [54]. Similar
targeting of PNAs to mitochondria also failed to show
inhibition of mtDNA replication in intact cells [21,23],
suggesting that access of the PNA to mtDNA in the
matrix was also limited.
In summary, even though an active DNA alkylating
reagent could be delivered to mitochondria there was no
evidence for its reaction with mtDNA in situ.These
unexpected findings suggest that the accessibility of
mtDNA to some alkylating reagents may be constrained.
It may be that more reactive alkylating reagents could be
used to modify mtDNA, but nonspecific reactions with
mtDNA or modification to nuclear DNA could limit this
approach. A better understanding of the selective alky-
lation of mtDNA in situ is required in order to develop
therapeutic strategies to deplete mutated mtDNA mole-
cules selectively.
2834 A. M. James et al. (Eur. J. Biochem. 270) Ó FEBS 2003
Acknowledgement
We thank P. Howard (School of Pharmacy, University of London) for
thegiftofDC81.
References
1. Chomyn, A., Cleeter, M.W.J., Ragan, C.I., Riley, M., Doolittle,
R.F. & Attardi, G. (1986) URF6, last unidentified reading frame
of human mtDNA, codes for an NADH dehydrogenase subunit.
Science 234, 614–618.
2. Wallace, D.C. (1999) Mitochondrial diseases in man and mouse.
Science 283, 1482–1488.
3. Schon, E.A. (2000) Mitochondrial genetics and disease. TIBS 25,
555–560.

4. Michikawa, Y., Mazzucchelli, F., Bresolin, N., Scarlato, G. &
Attardi, G. (1999) Aging-dependent large accumulation of point
mutations in the human mtDNA control region for replication.
Science 286, 774–779.
5. DiMauro, S. & Schon, E.A. (2001) Mitochondrial DNA muta-
tions in human disease. Am. J. Med. Genet. 106, 18–26.
6. Orth, M. & Schapira, A.H. (2001) Mitochondria and degenerative
disorders. Am.J.Med.Genet.106, 27–36.
7. Smeitink, J., van den Heuvel, L. & DiMauro, S. (2001) The
genetics and pathology of oxidative phosphorylation. Nature Rev.
Genet. 2, 342–352.
8. James, A.M. & Murphy, M.P. (2002) How mitochondrial damage
affects cell function. J. Biomed. Sci. 9, 475–487.
9. Chrzanowska-Lightowlers, Z.M.A., Lightowlers, R.N. & Turn-
bull, D.M. (1995) Gene therapy for mitochondrial DNA defects: is
it possible? Gene Ther. 2, 311–316.
10. Chinnery, P.F. & Turnbull, D.M. (1997) Mitochondrial medicine.
QJM. 90, 657–667.
11. Murphy, M.P. (1997) Targeting bioactive compounds to mito-
chondria. Trends Biotechnol. 15, 326–330.
12. Kolesnikova, O.A., Entelis, N.S., Mireau, H., Fox, T.D., Martin,
T.D. & Tarassov, I.A. (2000) Suppression of mutations in
mitochondrial DNA by tRNAs imported from the cytoplasm.
Science 289, 1931–1933.
13. Taylor, R.W., Chinnery, P.F., Turnbull, D.M. & Lightowlers,
R.N. (1997) Selective inhibition of mutant human mitochondrial
DNA replication in vitro by peptide nucleic acids. Nature Genet.
15, 212–215.
14. Taylor, R.W., Chinnery, P.F., Turnbull, D.M. & Lightowlers,
R.N. (2000) In-vitro genetic modification of mitochondrial func-

tion. Hum Reprod. 15, 79–85.
15. Taylor, R.W., Wardell, T.M., Lightowlers, R.N. & Turnbull,
D.M. (2000) Molecular basis for treatment of mitochondrial
myopathies. Neurol. Sci. 21, S909–S912.
16. Taylor, R.W., Wardell, T.M., Smith, P.M., Muratovska, A.,
Murphy, M.P., Turnbull, D.M. & Lightowlers, R.N. (2001) An
antigenomic strategy for treating heteroplasmic mtDNA dis-
orders. Adv. Drug Deliv. Rev. 49, 121–125.
17. Srivastava, S. & Moraes, C.T. (2001) Manipulating mitochondrial
DNA heteroplasmy by a mitochondrially targeted restriction
endonuclease. Hum. Mol. Genet. 10, 3093–3099.
18. Manfredi, G., Fu, J., Ojaimi, J., Sadlock, J.E., Kwong, J.Q., Guy,
J. & Schon, E.A. (2002) Rescue of a deficiency in ATP synthesis by
transfer of MTATP6, a mitochondrial DNA-encoded gene, to the
nucleus, Nature Genet. 30, 394–399.
19. Nielsen, P.E. (1999) Peptide nucleic acids as therapeutic agents.
Curr. Opin. Struct. Biol. 9, 353–357.
20. Taylor, R.W., Chinnery, P.F., Turnbull, D.M. & Lightowlers,
R.N. (1997) Selective inhibition of mutant human mitochondrial
DNA replication in vitro by peptide nucleotide acids. Nature
Genet. 15, 212–215.
21. Chinnery, P.F., Taylor, R.W., Diekert, K., Lill, R., Turnbull,
D.M. & Lightowlers, R.N. (1999) Peptide nucleic acid delivery to
human mitochondria. Gene Ther. 6, 1919–1928.
22. Muratovska,A.,Lightowlers,R.N.,Taylor,R.W.,Wilce,J.A.&
Murphy, M.P. (2001) Targeting large molecules to mitochondria.
Adv. Drug Deliv. Rev. 49, 189–198.
23. Muratovska, A., Lightowlers, R.N., Taylor, R.W., Turnbull,
D.M., Smith, R.A., Wilce, J.A., Martin, S.W. & Murphy, M.P.
(2001) Targeting peptide nucleic acid (PNA) oligomers to

mitochondria within cells by conjugation to lipophilic cations:
implications for mitochondrial DNA replication, expression and
disease. Nucl. Acids Res. 29, 1852–1863.
24. King, M.P. & Attardi, G. (1996) Isolation of human cell lines
lacking mitochondrial DNA. Methods Enzymol. 264, 304–313.
25. Puvvada, M.S., Hartley, J.A., Jenkins, T.C. & Thurston, D.E.
(1993) A quantitative assay to measure the relative DNA-binding
affinity of pyrrolo[2,1-c] [1,4]benzodiazepine (PBD) antitumour
antibiotics based on the inhibition of restriction endonuclease
BamHI. Nucleic Acids Res. 21, 3671–3675.
26. Hurley, L.H., Reck, T., Thurston, D.E., Langley, D.R., Holden,
K.G., Hertzberg, R.P., Hoover, J.R., Gallagher, G. Jr, Faucette,
L.F., Mong, S.M., et al. (1988) Pyrrolo[1,4]benzodiazepine anti-
tumor antibiotics: relationship of DNA alkylation and sequence
specificity to the biological activity of natural and synthetic com-
pounds. Chem. Res. Toxicol. 1, 258–268.
27. Petrusek, R.L., Anderson, G.L., Garner, T.F., Fannin, Q.L.,
Kaplan,D.J.,Zimmer,S.G.&Hurley,L.H.(1981)
Pyrrol[1,4]benzodiazepine antibiotics. Proposed structures and
characteristics of the in vitro deoxyribonucleic acid adducts of
anthramycin, tomaymycin, sibiromycin, neothramycins A and B.
Biochemistry 20, 1111–1119.
28. Hertzberg, R.P., Hecht, S.M., Reynolds, V.L., Molineux, I.J. &
Hurley, L.H. (1986) DNA sequence specificity of the
pyrrolo[1,4]benzodiazepine antitumor antibiotics. Methidium-
propyl-EDTA-iron (II) footprinting analysis of DNA binding
sites for anthramycin and related drugs. Biochemistry 25, 1249–
1258.
29. Puvvada, M.S., Forrow, S.A., Hartley, J.A., Stephenson, P.,
Gibson, I., Jenkins, T.C. & Thurston, D.E. (1997) Inhibition of

bacteriophage T7 RNA polymerase in vitro transcription by
DNA-binding pyrrolo[2.1-C][1,4]Benzodiazepines. Biochemistry
36, 2478–2484.
30. Burns, R.J., Smith, R.A. & Murphy, M.P. (1995) Synthesis and
characterization of thiobutyltriphenylphosphonium bromide, a
novel thiol reagent targeted to the mitochondrial matrix. Arch.
Biochem. Biophys. 322, 60–68.
31. Smith, R.A., Porteous, C.M., Coulter, C.V. & Murphy, M.P.
(1999) Selective targeting of an antioxidant to mitochondria. Eur.
J. Biochem. 263, 709–716.
32. Kelso,G.F.,Porteous,C.M.,Coulter,C.V.,Hughes,G.,Porte-
ous, W.K., Ledgerwood, E.C., Smith, R.A. & Murphy, M.P.
(2001) Selective targeting of a redox-active ubiquinone to
mitochondria within cells: antioxidant and antiapoptotic proper-
ties. J. Biol. Chem. 276, 4588–4596.
33. Murphy, M.P. (2001) Development of lipophilic cations as ther-
apies for disorders due to mitochondrial dysfunction. Expert Opin.
Biol. Ther. 1, 753–764.
34. Thurston, D.E., Murty, V.S., Langley, D.R. & Jones, G.B. (1990)
O-Debenzylation of a pyrrolo[2,1-c][1,4]benzodiazapine in the
presence of a carbinolamine functionality: Synthesis of DC-81.
Synthesis 81–84.
35. Lin, T K., Hughes, G., Muratovska, A., Blaikie, F.H., Brookes,
P.S.,Darley-Usmar,V.,Smith,R.A.&Murphy,M.P.(2002)
Specific modification of mitochondrial protein thiols in response
to oxidative stress: a proteomics approach. J. Biol. Chem. 277,
17048–17056.
Ó FEBS 2003 Mitochondrial DNA alkylation (Eur. J. Biochem. 270) 2835
36. Grossman, L.I., Watson, R. & Vinograd, J. (1973) The presence of
ribonucleotides in mature closed-circular mitochondrial DNA.

Proc.NatlAcad.Sci.USA70, 3339–3343.
37. Chappell, J.B. & Hansford, R.G. (1972) Preparation of
mitochondria from animal tissues and yeasts. In Subcellular
components. Preparation and Fractionation (Birnie, G.D., ed.), pp.
77–91. Butterworths, London.
38. Gornall, A.G., Bardawill, C.J. & David, M.M. (1949) Determi-
nation of serum protein by means of the biuret reaction. J. Biol.
Chem. 177, 751–766.
39. Kamo, N., Muratsugu, M., Hongoh, R. & Kobatake, Y. (1979)
Membrane potential of mitochondria measured with an electrode
sensitive to tetraphenyl phosphonium and relationship between
proton electrochemical potential and phosphorylation potential in
steady state. J. Membr. Biol. 49, 105–121.
40. Davey, G.P., Tipton, K.F. & Murphy, M.P. (1992) Uptake and
accumulation of 1-methyl-4-phenylpyridinium by rat liver
mitochondria measured using an ion-selective electrode. Biochem.
J. 288, 439–443.
41. James,A.M.,Wei,Y H.,Pang,C Y.&Murphy,M.P.(1996)
Altered mitochondrial function in fibroblasts containing MELAS
or MERRF mitochondrial DNA mutations. Biochem. J. 318,
401–407.
42. Brown, G.C. & Brand, M.D. (1985) Thermodynamic control of
electron flux through mitochondrial cytochrome bc1 complex.
Biochem. J. 225, 399–405.
43. Hafner, R.P., Nobes, C.D., McGown, A.D. & Brand, M.D.
(1988) Altered relationship between protonmotive force and
respiration rate in non-phosphorylating liver mitochondria isolated
from rats of different thyroid hormone status. Eur. J. Biochem.
178, 511–518.
44. Hafner, R.P., Brown, G.C. & Brand, M.D. (1990) Thyroid hor-

mone control of state-3 respiration in isolated rat liver
mitochondria. Biochem. J. 265, 731–734.
45. Brand, M.D. (1995) Measurement of Mitochondrial Protonmotive
Force in Bioenergetics – a Practical Approach (Brown, G.C., &
Cooper, C.E., eds), pp. 39–62. IRL, Oxford.
46. Wilson, S.C., Howard, P.W., Forrow, S.M., Hartley, J.A.,
Adams, L.J., Jenkins, T.C., Kelland, L.R. & Thurston, D.E.
(1999) Design, synthesis, and evaluation of a novel sequence-
selective epoxide-containing DNA cross-linking agent based on
the pyrrolo[2, 1-c][1,4]benzodiazepine system. J. Med. Chem. 42,
4028–4041.
47. Robin, E.D. & Wong, R. (1988) Mitochondrial DNA molecules
and virtual number of mitochondria per cell in mammalian cells.
J. Cell Physiol. 136, 507–513.
48. Porteous, W.K., James, A.M., Sheard, P.W., Porteous, C.M.,
Packer, M.A., Hyslop, S.J., Melton, J.V., Pang, C.Y., Wei, Y.H. &
Murphy, M.P. (1998) Bioenergetic consequences of accumulating
the common 4977-bp mitochondrial DNA deletion. Eur. J. Bio-
chem. 257, 192–201.
49. Munn, E.A. (1974) The Structure of Mitochondria.Academic
Press, New York.
50. MacFarlane, M., Merrison, W., Bratton, S.B. & Cohen, G.M.
(2002) Proteasome-mediated degradation of Smac during apop-
tosis: XIAP promotes Smac ubiquitination in vitro. J. Biol. Chem.
277, 36611–36616.
51. King, M.P. & Attardi, G. (1989) Human cells lacking mtDNA:
repopulation with exogenous mitochondria by complementation.
Science 246, 500–503.
52. King, M.P. & Attardi, G. (1988) Injection of mitochondria into
human cells leads to a rapid replacement of the endogenous

mitochondrial DNA. Cell 52, 811–819.
53. Ge
´
rard, B., Bourgeron, T., Chretien, D., Ro
¨
tig, A., Munnich, A.
& Rustin, P. (1993) Uridine preserves the expression of respiratory
enzyme deficiencies in cultured fibroblasts. Eur. J. Pediatr. 152,
270.
54.Takamatsu,C.,Umeda,S.,Ohsato,T.,Ohno,T.,Abe,Y.,
Fukuoh,A.,Shinagawa,H.,Hamasaki,N.&Kang,D.(2002)
Regulation of mitochondrial
D
-loops by transcription factor A
and single-stranded DNA-binding protein. EMBO Rep. 3,
451–456.
2836 A. M. James et al. (Eur. J. Biochem. 270) Ó FEBS 2003